© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 16570
    Prev
    Next
    INDEX
    Explanations

    closing quotation marks

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires strongly on tokens that begin a new sentence or quoted remark—i.e. sentence-start or quote-start words.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    他們
    -1.18
     diferentes
    -1.09
     different
    -0.93
    ncbi
    -0.89
     relevantes
    -0.89
     கே
    -0.88
     Letras
    -0.87
     rapidement
    -0.87
    Odkazy
    -0.86
     zukünf
    -0.85
    POSITIVE LOGITS
     I
    1.56
     we
    1.30
     my
    1.30
     since
    1.03
     ведь
    1.00
     ideally
    1.00
    ˆ
    0.98
     Ведь
    0.98
     berechnet
    0.97
    though
    0.96
    Activations Density 0.018%

    No Known Activations